cinera_handmade.network/cmuratori/hero/code/code068.hmml

36 lines
3.5 KiB
Plaintext
Raw Permalink Normal View History

[video output=day068 member=cmuratori stream_platform=twitch stream_username=handmade_hero project=code title="Exact Enforcement of Maximum Movement Distances" vod_platform=youtube id=0ZjYHfYAjQg annotator=schme annotator=jpike]
[0:28][Recap of where we are with collisions and interactions]
[1:07][Bug fix from forums - Typo in MoveEntity for testing entity flag]
[3:44][Overview for today - max movement distances]
[7:57][Review of MoveEntity]
[9:22][Review of tMin usage in MoveEntity]
[13:36][Starting implementation with a maximum distance]
[16:08][Discussion about cost of square roots]
[17:21][Using a ratio to cap tMin]
[21:49][Updating DistanceRemaining]
[22:41][TODO for Epsilon for (PlayerDeltaLength > 0) ?]
[23:37][Adding DistanceLimit to sim_entity]
[26:37][Decrementing DistanceLimit]
[28:07][Testing out the changes]
[28:55][Discussing collision response and double dispatch]
[42:55][Alternative to entity type double dispatch - fundamental quality double dispatch]
[47:26][Starting to add entity type double dispatch first for collision handling]
[54:30][Why sword-monstar collision won't work quite yet - need to respond to collisions but not stop movement]
[58:26][Q&A][:speech]
[59:40][How long in your opinion does it take to become fluent on Emacs?]
[1:00:10][Could an entity have multiple fundamental quality property things?]
[1:02:11][When are you planning to add depth sorting?]
[1:02:33][Will there be enough entities that some optimization will have to happen? How many do you expect to have on a screen? Hundreds?]
[1:04:29][I'm not sure I quite follow your argument for double-dispatch table versus fundamental qualities. Seems to me like you still need to figure out what to do for each quality, what to do in the presence of others, still leads to quadratic growth.]
[1:10:54][Yesterday you were talking about using callbacks for the collision system. Will this new system replace the need for them?]
[1:11:30][Casey, I asked a question and the Twitch app crashed on iPad. Any chance you my questions before it did?]
[1:12:20][Casey, as a casual observer and occasional troll, where is the game now? Seriously, I haven't watched for a while, and how do you find the energy to do this weeknights?]
[1:16:34][I noticed sometimes you try to determine regions for collision boxes. Why not implement basic shapes and use those as debug helpers with some alpha percentage?]
[1:18:16][Do you test for collision every update interval? If your sword overlaps with an entity, could it happen that a moving entity gets missed because both entity and sword move in the same tick above and beyond each other?]
[1:20:27][To me it seems as the sword should be influenced by the speed of the player movement at the time it was dispatched? Am I right?]
[1:23:24][I missed the part on how many entities you are simulating at once and where you store them, but won't the dispatch table approach be negative for cache performance? Could its simulation be deferred?]
[1:24:58][It seems logical to make infinity the default maximum distance an entity can travel, but that could have performance implications. Could you say something about that?]
[1:26:14][Could you also do something like that for AI, where for instance, you can impose simple behaviors on top of each other randomly, to make complex and varied behavior, instead of doing the more common behavior tree approach?]
[1:26:32][If you do the AAA-ness way, can you come back to it later and edit or add more details to it like from 6-to-8, etc., or will that cause problems?]
[/video]